Artificial Intelligence and machine learning are increasingly used to interrogate complex biological data. This systematic review evaluates their application to multi-omics for the molecular characterization of hematological malignancies, an area with unmet clinical need. We searched PubMed, Embase, Institute of Electrical and Electronics Engineers Xplore, and Web of Science from January 2015 to December 2024. Two reviewers screened records, extracted data, and used a modified appraisal emphasizing explainability, performance, reproducibility, and ethics. From 2847 records, 89 studies met inclusion criteria. Studies focused on acute myeloid leukemia (34), acute lymphoblastic leukemia (23), and multiple myeloma (18). Other hematological diseases were less frequently studied. Methods included Support Vector Machines, Random Forests, and deep learning (28, 25, and 24 studies). Multi-omics integration was reported in 23 studies. External validation occurred in 31 studies, and explainability in 19. The median diagnostic area under the curve was 0.87 (interquartile range 0.81 to 0.94); deep learning reached 0.91 but offered the least explainability. Artificial Intelligence and machine learning show promise for molecular characterization, yet gaps in validation, interpretability, and standardization remain. Priorities include external validation, interpretable modeling, harmonized evaluation, and standardized reporting with shared benchmarks to enable safe, reproducible clinical translation.

Duchenne muscular dystrophy (DMD) is a fatal X-linked disorder, characterized by progressive skeletal muscle wasting. The disease is caused by various types of mutations in the dystrophin gene (DMD). The disease occurs at a frequency of about 1 in 5000 male births, making it the most common severe neuro-muscular disease. In addition to clinical examinations of muscle strength and function, diagnosis of DMD usually involves a combination of immunological assays using muscle biopsies, typically immunohistochemistry and western blotting, and molecular techniques such as DMD gene sequencing or Multiplex Ligation Dependent Probe Amplification (MLPA) using blood samples. In fact, precise molecular diagnosis is a prerequisite for determining the appropriate personalized therapeutic approach such as exon-skipping, gene therapy or stem cell-based therapies in conjunction with gene editing techniques like CRISPR-Cas9. However, the quest for reliable biomarkers with high sensitivity and specificity for DMD from liquid biopsy is still a hotspot of research, as such non-invasive biomarker(s) would not only facilitate disease diagnosis but would also help in carrier detection, which will eventually result in better disease management. In this chapter, we will illustrate the detailed current and prospect strategies for disease.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel coronavirus that has been identified as the causative agent of an ongoing global pandemic of acute respiratory distress syndrome, coronavirus disease 2019. Ranked as one of the top priorities is developing an accurate, reliable, and affordable laboratory diagnostic test for pathogen detection to facilitate disease tracking. In this study, a multiplex-based assay to detect SARS-CoV-2 using four primers/probes to amplify regions of the spike (S), nucleocapsid (N), and envelope (E) protein-coding genes, along with amplifying region of the 3' untranslated region (UTR) was designed and developed. To provide accurate detection of different SARS-CoV-2 variants, an in silico design and validation pipeline was utilized. These included multi-sequence alignments with Clustal Omega, specificity verification by the Basic Local Alignment Search Tool (primer-BLAST), and computational performance verification using Biopython. The primers and hydrolysis probes designed were experimentally tested against more than two million SARS-CoV-2 genome sequences obtained from the National Center for Biotechnology Information Virus Database, with >97% specificity across 2200 lineages, including all of the significant variants of concern and variants of interest. The experimental validation phase included synthetic SARS-CoV-2 constructs and ribonucleic acid samples from nasopharyngeal swabs of 20 patients, including both symptomatic and asymptomatic cases. The assay displayed 100% sensitivity and specificity, with the limit of detection as low as 10 copies/mL, which surpassed the detection limits of some commercially available reverse transcription quantitative polymerase chain reaction (RT-qPCR) kits. Amplification efficiency was optimized for all targets, ranging from 94.92% for the N gene, 98.84% for the S gene, 81.75% for the E gene, to 82.14% for the 3′UTR genes. Assay selectivity was verified by cross-reactivity against common respiratory viruses—SARS-CoV-1, Middle East respiratory syndrome coronavirus, Influenza A/B, and human coronaviruses (HCoV-229E, HCoV-NL63, HCoV-OC43, and HCoV-HKU1)—and no off-target signal was observed. The ability to detect multiple genomic regions renders the assay resilient to mutation-based primer mismatch, making it extremely suitable for clinical diagnostics, epidemiological surveillance, and variant monitoring.